The overall objectives of this project are to study the phenomenology and biological substrates of changing responsivity to drugs. These findings are used to develop hypotheses about clinical loss of response to the anticonvulsant mood stabilizers and to test possible ways of showing or reversing the tolerance process. The models under study also have the unique feature of being dependent on the contingencies or context of drug administration. As such, they do not involve primary changes in pharmacokinetics, but instead depend upon adaptations in the nervous system in response to the drugs as they interact with illness state. In the kindling paradigm, we measure the loss and reinstatement of anticonvulsant drug efficacy, and in the sensitization model, we study the environmental context determinants of drug response. We have demonstrated that the anticonvulsant efficacy of carbamazepine, valproate, and diazepam can be manipulated by temporal factors relating to drug administration and seizure presentation. Contingent changes in physiological responsivity, gene expression, and receptor regulation have been demonstrated in concert with the development of tolerance. For studying cocaine sensitization,we have developed a rapid behavioral paradigm that produces a conditioned or context-dependent enhanced behavioral response to cocaine; the neurobiological correlates of this continue to be investigated as they relate to our understanding of how environmental cues can be associated with changes in behavior (i.e., learning). Significant findings to date include demonstration of the following. 1) Contingent inefficacy and tolerance to carbamazepine and other anticonvulsants, whereby the contingent presentation of the drug before, but not after, kindling stimulation results in a diminished response to the drug in various stages of kindling evolution. 2) Reversibility of contingent tolerance by time off treatment or even continued treatment with the drugs given after the kindled seizures (i.e., drug administration is not discontinued, only the contingencies are changed) or by kindled seizures alone. 3) Selective cross tolerance between carbamazepine and lamotrigine, valproate, and PK-11195 (an antagonist at the peripheral-type benzodiazepine receptor), but not diazepam, clonazepam, or phenytoin. 4) Alterations in seizure threshold which mirror the changes in responsivity to carbamazepine. 5) Procedures to slow contingent tolerance development; e.g., noncontingent drug presentation or kindling the rats at lower stimulation currents, but not co- administration of the NMDA antagonist MK-801 or the calcium channel blocker nimodipine. 6) A number of neurochemical correlates of contingent tolerance that represent a loss of a subset of seizure-induced adaptations in the GABA-A system, as well as in peptide mRNAs, trophic factors, and immediate early genes. 7) The failure of TRH mRNA to show seizure-induced increases in carbamazepine-tolerant animals has been more closely mechanistically linked to tolerance, with the observation that TRH injected bilaterally into the hippocampus is anticonvulsant and also increases the effectiveness of carbamazepine in tolerant animals. 8) Neurochemical correlates of contingent tolerance to diazepam which are highly similar to those observed in carbamazepine tolerance despite the differential mechanisms of action of the two anticonvulsants. 9) Tolerance to both carbamazepine and diazepam has been associated with failure of seizures to increase the mRNA specificity for the alpha-4 subunit of the GABA receptor. 10) Slower tolerance development to combined treatment with carbamazepine (15 mg/kg) and valproate (low dose [150 mg/kg]) compared with either one alone. 11) Oscillatory patterns of drug responsivity to carbamazepine and valproate under certain circumstances of repeated drug administration and kindling stimulation (i.e., minimally effective doses or lower stimulation intensities). 12) Development of a novel one-day cocaine sensitization paradigm that is entirely conditioned or context-dependent and dependent on an intact amygdala and nucleus accumbens. 13) The requirement of intact dopamine function for the development, but not expression, of context- dependent sensitization. 14) Cross sensitization between cocaine and the NMDA antagonist MK-801, and between cocaine and procaine (a local anesthetic that is also self-administered by primates), but not between cocaine and lidocaine (which is not self-administered) or cocaine and caffeine. 15) Blockade of the development of sensitization by MK-801, lithium, nimodipine, clonidine, and diazepam, but not by carbamazepine, proglumide (CCK antagonist), or alpha-helical CRF (CRF antagonist). 16) Blockade of the expression of conditioned sensitization by clonidine, diazepam, and nimodipine.